diff --git a/llama_stack/distribution/server/server.py b/llama_stack/distribution/server/server.py index 6a0047f69..bb9ef0361 100644 --- a/llama_stack/distribution/server/server.py +++ b/llama_stack/distribution/server/server.py @@ -342,7 +342,8 @@ def main(): app = FastAPI(lifespan=lifespan) app.add_middleware(TracingMiddleware) - app.add_middleware(ClientVersionMiddleware) + if not os.environ.get("LLAMA_STACK_DISABLE_VERSION_CHECK"): + app.add_middleware(ClientVersionMiddleware) try: impls = asyncio.run(construct_stack(config))